326 



which all the members of the Oleaceae, to which they, together with 

 the lilac, the olive, and others belong, may be grafted on each 

 other ; " a fact," as Von Martins observes, " which demonstrates the 

 analogy of their juices and their fibres." Even the jasmine, so like 

 the Oleaceae in many respects, will not unite by grafting with any of 

 them. The association of " the potato with capsicum and the deadly 

 nightshade " is equally natural, whether we regard their structure or 

 their properties. The potato {Sola/turn tuberosutn) belongs to the 

 same genus as the poisonous nightshades (>S'. nigrum and »S'. Dulca- 

 mara), and participates in their deleterious qualities to a considerable 

 extent. The leaves and berries of the potato are poisonous, as are 

 those of the nightshades ; the underground stems or tubers in a raw 

 state also partake of the same poisonous nature, and are rendered fit 

 for food only by cooking. The same may be said of the other mem- 

 bers of the order ; they are all more or less of a poisonous nature. 

 Indeed, as DeCandolle has well observed, " We must not lose sight 

 of the fact, that all our aliments contain a small quantity of an excit- 

 ing principle, which is necessary as a natural condiment, but which 

 in excess would be injurious." Nor do we see what objection can 

 be made to the junction of the Veronica with Verbascum and the 

 other Scrophulariaceae, except on the score of its having only two 

 stamens : this character is uncertain, as we have before shown ; but 

 that of insertion on the corolla is certain, as is that of the capsule 

 being two-celled and many-seeded. 



The quotation in the ' Observations ' from the ' Vegetable King- 

 dom,' containing the characters of the order Caprifoliaceae, only proves 

 the author's inability to appreciate the advantages of the structural 

 system. When all the parts of the plant are looked to in drawing up 

 the character of an order, that character must necessarily be so com- 

 prehensively framed as to meet all the cases of variation that may 

 occur in the non-essential organs ; but then there are other organs 

 which furnish the certain character, and these do not vary. For 

 example, the Caprifoliaceae belong to an alliance — the Cinchonales — 

 the members of which all agree in the possession of an inferior ovary, 

 flowers having both calyx and corolla, the corolla monopetalous, and 

 a minute embryo lying in a large quantity of albumen. The order 

 itself is characterized by its epipetalous stamens, equal in number to 

 the lobes of the corolla, and alternate with them, the anthers straight, 

 and opening longitudinally for the discharge of the pollen, a consoli- 

 dated fruit, and leaves without stipules. One would have thought 



